In my guest blogging gig for MASNSports.com today, I wrote about Bryce Harper’s eighth inning sacrifice, Win Probability Added, and human evolution. It was a bit of a rambler, but my biggest point was this:
You don’t get to pick and choose which stats you think are the right ones. They all are.
It drives me absolutely crazy to hear fans, players, managers or executives dismiss certain statistical evaluators, like we’re fabricating these numbers or pulling them out of thin air. WAR, or WPA, or wOBA, or wRC+, or ISO, or FIP, or UZR… all of those numbers are in the game every bit as much as batting average or earned run average.
It’s just that those “in the game” have been using the traditional statistical evaluators for over a century and some others were “invented” by folks not actually “in the game” in the past two decades.
Just because a statistical evaluator was created by math whiz doesn’t mean it’s any more or less legitimate than those we’ve been using for 120 years.
Each, in their own way, tells part of the story about what’s going on out there. No single statistical evaluator can tell us exactly how efficient a particular player is in his chosen craft. Some of them give us a better idea than others. But each should be taken in the context it is presented.
The “new stats” weren’t created to make following the game more difficult. They were developed to help us more deeply understand the game. Or help us compare players on a more neutral field. Or help us compare current players against the past more accurately. They weren’t created to confuse, but enlighten.
Fangraphs.com has a glossary of many of the “new stats”. They don’t hide their formulas. There’s a lot to take in, but if you take a couple of minutes most of the “new stats” are pretty simple to understand. Sure, there are some concepts that might take a few moments to think about before they make total sense. But they are all as rooted in the game as ERA, which is not a particularly good or accurate method to evaluate a pitcher.
Here’s another chunk of my MASN column to think about:
Back in the old days, they invented batting average and earned run average as a method of evaluating players side-by-side since they weren’t able to watch every game in person.
Yes, there was an era before computers. Before television. Even before radio was popular. If you wanted to know what type of ball player a guy was, you has to see him in person. You had to travel for days and hope for no rain out. There was little scouting and even less statistical evaluation. That’s why they started to keep track of these things, in order to be able to evaluate players without actually seeing them in person.
Even though every single game is now on TV and we have video of each player going back to their middles school games, we’re still looking for more clear statistical evidence to measure a player’s effectiveness. PITCHF/x and batted ball data are taking us into the next phase of statistical evaluation, and it all helps us better understand the game.
Statheads and seamheads have been at odds for decades. They don’t have to be. Each individual statistical evaluator only tells part of the story. Taken in context, they are part of the big picture. If you love the game, it’s worth your while to become more familiar with these concepts. It’s just a little math, that’s all.